© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 1-RES-SM
    11. 25730
    Prev
    Next
    INDEX
    Explanations

    the abbreviation "IL" indicating a specific state or context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/1-res-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.1.hook_resid_post
    Hook Layer
    1
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    394
    +0.14
    0.8%
    410
    +0.12
    0.7%
    81
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    394
    +0.14
    0.02
    285
    +0.12
    0.01
    378
    +0.10
    0.02
    Negative Logits
    fections
    -1.98
    Ĺ
    -1.57
    gmail
    -1.49
    fection
    -1.48
    jury
    -1.45
     finding
    -1.42
     sudden
    -1.41
     injuries
    -1.39
     ')
    -1.38
     true
    -1.37
    POSITIVE LOGITS
    lement
    2.42
    inux
    2.25
    ungen
    1.91
    ensed
    1.89
    INGS
    1.80
    ander
    1.80
    ibrary
    1.76
    oque
    1.72
    LECT
    1.72
    ensing
    1.71
    Activations Density 0.058%

    No Known Activations